Oi, Andy

You reached for the one tax that would actually work. Here is the operating manual nobody handed you, and the reason you will probably never be allowed to use it.

You won. And before the votes were even counted, two newspapers, the chairmen of two parties and a former head of the farmers' union were already attacking a tax. Not a tax you have introduced. A tax you might introduce, if you take the leadership, if you remove the Prime Minister first. Three conditionals deep, and the alarm was already deafening.

That reaction is not noise. It is the most useful piece of information you will be handed this year. Nobody clears a front page to stop a policy that would not bite. The volume is proportional to the threat, and the threat is real, because for once a frontline politician has pointed at the actual problem. So this is written in your favour, mostly. You have reached for the one lever that reaches the disease.

The bad news comes in two parts. The first part you can fix. The second part you cannot, and I am going to be straight with you about which is which.

The wall is made of collateral

The objection that matters does not come from the Tory chairman or the farming lobby. It comes from the people who agree with you. Charles Goodhart spent a career at the Bank of England and thinks the case for a land tax is overwhelming. He will also tell you why it cannot be done in one move. Bring it in suddenly and land values fall, not a little, a great deal. And because the entire credit system is built on property as collateral, a fall like that does not stay inside the housing market. It becomes a banking event.

The reason you cannot tax the land is that the country has been wired on top of it. The tax base is holding up the building.

This is not a thought experiment. Britain ran it. In 1909 a government you would recognise as your ancestors passed taxes on land value. The taxes devalued building land. That wrecked the collateral housebuilders borrowed against. Construction collapsed. Within a decade the duties were quietly repealed and filed under valuation difficulty. Same asset, same collateral, same retreat, one hundred and seventeen years ago. So when the property industry says it always fails, they are not lying to you. They are describing a trap.

The only question worth asking is whether the trap has an exit. It does. Here it is, in six moves.

---

The way through, in six moves

One. Phase it in, and stand behind the banks with real money.

Start low and climb. Half a percent rising to two and a half over five years. But be clear with yourself about what the phase-in does and does not do, because this is where the well-meaning versions of your policy go to die.

Markets are forward-looking. The day you credibly announce a terminal rate, the market prices the whole future stream and reprices the asset at once. The phase-in smooths the cash a landlord has to find each year, which matters for politics and for pass-through, but it does not smooth the hit to collateral, which is the thing Goodhart is actually worried about. What defuses the credit event is not the gentle rate. It is the work you do on the banking system before a single pound of the new tax is ever collected.

So start two years out, quietly, with the Bank of England, and turn the dials the other way from where they have been jammed since 2008. Affordability tests go up, not down. Maximum loan-to-value on new lending comes down. The capital that banks must hold against their mortgage books goes up, so the system is carrying a thicker cushion by the time the tax arrives. The reason is brutally simple. You do not want to wake up on implementation day with a banking system whose mortgage book is rammed full of week-old ninety per cent loans, written against valuations your own policy is about to lower. You want the freshest, thinnest-equity lending to have stopped well before the repricing, so that the loans most exposed to negative equity are old enough to have paid down and inflated away some of the gap.

Then the backstop itself, and be honest about its scale, because the number frightens people and it should not. The country's outstanding mortgage debt is around 1.7 trillion pounds, spread across roughly twelve and a half million accounts. You are not insuring that. Most of it sits at comfortable loan-to-value ratios and never goes near negative equity even under a serious markdown. What you are insuring is the tail: the slice of recent, high-leverage lending that a fall in land values would push underwater, together with the short-term cash the banks need so that a markdown never forces a fire sale. That is a contingent facility in the tens of billions, committed and visible from day one, not spent. It is closer in spirit to the guarantees thrown around the system in 2008 than to a cheque anyone actually writes, and like those guarantees its real job is to be believed, so that the panic never starts. The precise figure, the true size of that tail under a given repricing path, is the single most important number in the entire programme, and it is the one thing I would model to death before saying another word about this in public. Get it right and the wall comes down. Underfund it and Goodhart wins.

---

Two. Tax the rent, not the valuer's guess.

The thing that killed 1909 was valuation. You cannot cheaply and defensibly value the land under every building in the country, and the moment you try, the lawyers bleed you to death in the courts. So do not try. And here is the elegant part, the part I would build the whole reform around: you do not need a new army of valuers, because the number you want already exists. Every landlord in the country already tells HMRC what their property earns, in rent, every year, on a self-assessment return. So take it. Take the declared rent, capitalise it into a value with a standard multiple, and run that value through the bands. The rental market values itself, every year, automatically, out of data the state already collects. No surveyors, no tribunals, no reopening of 1909. You simply make the owner's own declaration the basis of the tax.

Then borrow the oldest anti-cheating device in European fiscal history. For four centuries Denmark ran the Sound Toll. Every ship passing the strait declared the value of its cargo, and the Danish crown reserved the right to buy the cargo at the declared price. Lie low and you lose the cargo. Apply the same logic here. You self-declare the rent. If you lowball it, the state, or a buyer, can take the property at the price your declaration implies. Modern economists have a name for this, the self-assessed tax with a purchase option, and a literature to go with it, so you will not be accused of inventing it on the back of an envelope.

One rule is not negotiable. The purchase option applies only to let and investment property, never to anyone's home. Forced sale of a family home is intolerable, and you should never let it within a mile of the proposal.

Now make the rent itself the mechanism. Set the tax in bands, and ratchet them so that raising the rent pushes the landlord into a higher band and a bigger bill, while lowering the rent drops him into a lower one. Done properly, putting the rent up costs the landlord more in tax than he gains in rent. That is your answer to the oldest objection of all, the one that says the tax just gets passed to tenants. Under the ratchet, passing it on makes the landlord worse off, and cutting the rent makes him better off. Raising rents makes him much worse off on the bill. Lowering them makes him much better off. The incentive is welded the right way round for the first time in living memory. Two birds, one stone: the same mechanism that values the property also makes cutting the rent the landlord's most profitable move, which is the nearest thing to an automatic brake on pass-through anyone has designed.

This is not a fantasy mechanism. Switzerland runs the nearest thing to it and has one of the most stable rental sectors in Europe to show for it. Swiss rents are tied by law to a published reference interest rate and surrounded by tenant protections, so that rents move slowly and predictably, and can even be pushed down by tenants when underlying costs fall, in a country where most people rent for the long term and the market still turns over and still builds. The lesson is not that regulation freezes a market. It is that a rental market which has been told, in effect, that the rent is the thing being watched can stay both liquid and calm. The part worth borrowing is the rental side, where pinning the tax to the rent has sat alongside stability for decades.

The catch, stated plainly, this only works if the bands are fine enough that there is no slack to raise rent inside a band, and not so fine that you have quietly reinvented rent control, which would shrink the supply you cannot afford to lose. The reconciliation is move three, plus a public builder holding the floor underneath the whole market.

And now close the dodge that every property tax invites: the empty flat. If the tax is assessed on declared rent, the obvious move is to declare no rent at all, leave the place dark, and wait. So the empty property is exactly where the tax has to be at its most aggressive, never its most forgiving.

Here is the process. A property with no declared tenancy and no owner living in it is, by definition, being withheld from use. Allow a short grace period for genuine voids, the months between tenants, a probate, a real renovation, call it six months. After that, the unit is no longer assessed on a rent it is not earning. It is assessed on its capital value directly, and the land tax on it escalates the longer it stays empty. Standard charge once it crosses the grace line. Double after a year dark. Treble after two. And it keeps climbing from there, because the whole point is that there is no level of vacancy at which holding an empty asset is cheaper than putting it to use. The valuation for an empty unit is not a fresh survey to be argued over, either; it is the last known assessed value, uprated, so refusing to let the place creates no gap to hide in.

The friend-for-a-pound move dies here too, against the purchase option from a moment ago. Declare a peppercorn rent to slip under the escalator, and you have told the state the flat is worth a peppercorn, at which price someone is now entitled to buy it from you. The empty-homes problem that every council in the country has flailed at for a generation turns out to dissolve the instant leaving a property empty costs more than filling it.

---

Three. Make planning permission a clock, not a trophy.

The standard complaint, that a land tax still will not get anything built, is fixable by aiming it. Grant permission and the clock starts, with an agreed delivery window inside which there is no land tax to pay at all. Build it out on time and you never pay a penny on the site. That is the reward for doing the thing the country actually needs done.

Miss the window and the meter starts, and it runs harder the longer you sit. Run past schedule and you pay the full land tax on the land. Twelve months late and it rises to one and a half times. Twenty-four months late and it is three times the standard charge, and still climbing from there. Sitting on permitted land, which has been the most profitable passive position in British development for forty years, becomes the single most expensive thing you can do with a plot.

The fairness valve is the clock-stop. If the delay is demonstrably not yours, the grid that will not connect you, the judicial review that freezes the site, the supplier who failed to deliver, you make that case and the escalator pauses. But if you are simply slow, that is your cost to carry. And if you are fast, the saving is your profit to keep. The incentive points one way only: build, and build promptly. Nobody is forcing a spade into the ground at a loss. The tax just ends the long era in which doing nothing with permitted land was the smartest play on the board.

---

Four. Tax the building, but price the land.

A pure land value tax needs you to separate land value from the bricks on top of it, which is the hard, litigable part, and the part that buried Lloyd George. Skip it. Tax the assessed property value, which you already have, and apply a density multiplier that recovers the land share.

A detached house on a large plot is almost all land, so it carries a multiplier of one. A mid-rise block spreads its plot across many homes, so each home carries a fraction, perhaps a fifth. A tower spreads it across hundreds, so a twentieth. You recover the Georgist incentive, build densely and stop hoarding land, without ever conducting a formal land valuation. Be honest that this approximates a land tax rather than being one. That honesty is a strength. It heads off the purists who will tell you it is not real Georgism, and it keeps you out of the courtroom that ended the last attempt.

There is one trap inside it. A low multiplier on a tower means the empty investment flat in a prime block, exactly the parked capital you most want to reach, is barely touched. Which is why move four only works bolted to move two. State the rule as a single sentence and it comes clean: the multiplier rewards occupied density, real homes with real people in them, and the vacancy escalator punishes empty density, capital parked in the sky. Reward the first. Hammer the second.

---

Five. Protect the pensioner, then phase her in.

The emotional core of every attack on you is the widow in the family home she bought for very little and cannot afford a new tax on. Take her off the board at the start. Exempt primary residences up to two million, which draws the teeth from the transition while still raising serious money from investment property and the genuinely expensive. Then bring that cap down slowly across a decade, so owner-occupiers are phased in gently rather than shocked in a single budget.

But do better than simply deferring her bill until she dies, because a tax whose central promise is that it can be paid from beyond the grave is grim, and worse, it leaves her rattling around a four-bedroom house that could hold a family. Give her a reason to move instead. Offer a capital gains tax holiday on the way out: an owner over a certain age who sells the large home and buys something smaller pays no tax on the gain and keeps the difference in cash. The tax stops being a threat and becomes a nudge, and the nudge releases precisely the under-occupied family housing that no amount of new building ever seems to free. For the genuinely asset-rich and income-poor who simply will not move, deferral stays on the table as a backstop. But the better outcome, for her and for the housing stock alike, is the door held open, not the bill rolled forward.

The same logic rescues a group nobody mentions: the people who bought one or two flats to let because a pension never looked like enough, and who are now sitting on a retirement made entirely of the asset you are taxing. Hit them with the full apparatus on day one and you get a political disaster and a fire sale at once. Offer them the same exit, a time-limited capital gains holiday to rotate out of the buy-to-lets across the transition, and you convert a cornered constituency into willing sellers, which is exactly the supply you want reaching first-time buyers. Tax a man into a corner and he fights you to the death. Give him a costed, time-limited way out, and a surprising number simply take it. Keep both holidays inside the transition window and capped, so they stay a bridge out of the old regime rather than a permanent loophole into it.

Here I owe you the part that matters most, because it is the part your cheering section will skip. The decaying exemption is exactly the kind of concession that the structure you are about to enter freezes in place. The moment the cap drops far enough to touch the median homeowner who votes, the pressure to stop it right there will be overwhelming, and the temporary exemption becomes permanent. The same framework that says this tax would work also predicts that this particular sweetener never fully dissolves, which quietly smuggles back the distortion you were trying to remove. I am not telling you to drop the exemption. You cannot pass the thing without it. I am telling you to expect it to get stuck, because the same machine that is attacking you this week is built to make it stick.

---

Six. Give the money somewhere to go.

This is the move your advisers will skip, and it is the one that makes the other five worth doing. A land tax on its own simply lowers the price of land. Useful, and incomplete. The point was never to punish land. It was to move capital off it and into things that actually produce.

So the tax has to run alongside a destination. A real national investment bank with the scale and the mandate to deploy at industrial scale. Zero capital gains tax on productive investment that stays in the country. An energy programme that stops British industry paying half as much again for electricity as its German competitors. Dislodge the capital with one hand and offer it a better home with the other. Skip this and you get the pain without the point, a cheaper housing market bolted to the same stagnant economy.

Stare at one timing problem honestly. The pain from the tax lands in the first five years. The payoff from the investment compounds over ten and beyond. There is a valley in the middle where the cost has arrived and the benefit has not. The backstop and the exemption are, among other things, what carry you across that valley alive.

---

Why this probably fails anyway, and what that tells you

So there it is. The wall the property industry swears is permanent is, on the economics, nothing of the sort. Phase it, backstop it, tax the rent and not the valuer, make permission a clock, price land through density, protect the pensioner and then phase her in, and give the freed capital a productive home. The collateral problem is soluble. The pass-through problem is soluble. The valuation problem dissolves. The supply problem is fixable. None of that is the reason you will struggle.

You will struggle because the design moves the impossibility from one place to another, and the new place is harder. It is no longer the economics that forbid the tax. It is the politics and the public finances, and both of those obstacles fall straight out of the same structure that produced the problem in the first place. A tax that disturbs the existing stock of wealth has to be carried against the people who hold that stock, and they are organised, and they vote, and a few of them write the front pages. A tax on a mere flow can be passed against a diffuse and disorganised public who barely notice. The first is always harder than the second, in every economy shaped like ours. That is not cynicism. It is a property of the configuration, and you can write it down as cleanly as any of the moves above.

Which is the part nobody around you will say, Andy, so I will. The reaction that greeted you this week, before you had even won the seat, is not an obstacle on the road to the policy. It is the policy's natural predator, and it was already fully grown and lying in wait. The exit exists. It is costed. It is built. The reason no one has ever been allowed to walk through it is the thing you watched defend itself across the front pages this week, and it will still be standing there long after the leadership is settled.

The manual is yours. The full costed version is on the common platform. The economics is the easy part, and I have just handed it to you. The politics is your problem, and mine, and everyone's who has ever rented a flat they could not buy.

Good luck. Whatever else you do, fund the backstop.